We have successfully identified a number of novel classes of natural products produced by marine organisms with very potent activity against murine and human tumors. Specifically, in the last grant period more than twenty papers have been published describing new biologically active metabolites. Major advances have also been made in the level of sophistication and significance of the biological and the pharmacological components of the program. The result has been the submission of two patents and three manuscripts describing the pharmacology of the compounds isolated as part of this program. Although the overall objectives of the program remain unchanged (to characterize potential antitumor metabolites produced by marine invertebrate animals), the approaches to accomplishing this goal have changed greatly. One of the lessons that we have learned during this last grant period is that cytotoxicity data against a single cell line is inadequate to judge the potential utility of new classes of metabolites and to prioritize them for further study. Consequently, the sub-contract for L1210 screening at Vermont was terminated and a new program was established at Utah. Initial screening of extracts for antineoplastic activity is now performed against a panel of human solid tumor cell lines. This panel includes several cell lines with specific resistance phenotypes that were developed here at the University of Utah and a panel developed at Lederle Laboratories. Partially purified fractions and pure compounds are tested in a second panel of drug sensitive Chinese Hamster Ovary (CHO) mutant cell lines that provide information about the mechanism of action, in vitro enzyme inhibition assays, and finally in vivo studies in athymic mice against the specific human tumor(s) that display sensitivity in vitro. The mechanism of action studies and enzyme inhibition assays focus primarily on interaction with, and inhibition of DNA function. The CHO mutants display specific deficiencies in DNA repair of single strand breaks, double strand breaks, or formation of bulky adducts. These assays are complemented by topoisomerase (TOPO) I and II inhibition assays, DNA strand cleavage and ethidium bromide exclusion assays. Although there is a clear emphasis placed in the proposal on metabolites that interact with DNA, metabolite, with promising activity but other mechanisms of action will not be excluded or selected against by the initial screening protocol. Bistramide A is an excellent example. It did not show activity in the DNA oriented mechanism screens, however it did show a very promising profile in the cell based screen at Lederle. Based on these results, bistramide A was evaluated in vivo and showed significant activity against the ovarian carcinoma Ovcar3, and in fact was more active against the solid tumor than against P388. In fact we believe that our protocol is capable of selecting for compounds active against human solid tumor diseases.